Neurol India Home 
 

CASE REPORT
Year : 2004  |  Volume : 52  |  Issue : 1  |  Page : 111--115

Normal perfusion pressure breakthrough in arteriovenous malformation surgery: The concept revisited with a case report

S Kumar, Y Kato, H Sano, S Imizu, S Nagahisa, T Kanno 
 Department of Neurosurgery, Fujita Health University, 1-98 Dengakubakubo, Kutsukake-cho, Toyoake, Aichi-470-1192, Japan

Correspondence Address:
Y Kato
Department of Neurosurgery, Fujita Health University, 1-98 Dengakubakubo, Kutsukake-cho, Toyoake, Aichi- 470-1192
Japan

Abstract

The pathophysiological mechanisms and the salient features of normal perfusion pressure breakthrough (NPPB) are discussed on the basis of an operated case of arteriovenous malformation.



How to cite this article:
Kumar S, Kato Y, Sano H, Imizu S, Nagahisa S, Kanno T. Normal perfusion pressure breakthrough in arteriovenous malformation surgery: The concept revisited with a case report.Neurol India 2004;52:111-115


How to cite this URL:
Kumar S, Kato Y, Sano H, Imizu S, Nagahisa S, Kanno T. Normal perfusion pressure breakthrough in arteriovenous malformation surgery: The concept revisited with a case report. Neurol India [serial online] 2004 [cited 2021 Sep 23 ];52:111-115
Available from: https://www.neurologyindia.com/text.asp?2004/52/1/111/6717


Full Text

 Introduction



Multifocal hemorrhage along with cerebral edema can be a frightening and sometimes catastrophic postoperative complication in high flow cerebral arteriovenous malformation (AVM) surgery. This phenomenon, termed as 'normal perfusion pressure breakthrough' (NPPB), has been attributed to the diversion of bloodflow from the AVM, after its resection, into the adjacent cortical blood vessels. A case of arteriovenous malformation is presented and the changes in the hemodynamics during and after surgery are assessed.

 Case Report



A 44-year-old male presented with frequent episodes of generalized seizures. There was no neurological deficit. Investigations revealed arteriovenous malformation in the left parieto-occipital lobe. Left carotid and vertebral angiograms showed large high flow feeders to the AVM from the posterior cerebral, middle cerebral as well as anterior cerebral arteries. The venous phase delineated the draining veins to the superior sagittal sinus as well as to the internal cerebral veins [Figure:1].

C-99M Single photon emission computed tomography (SPECT) showed a hypervolemic lesion adjacent to the nidus. In 123I-IMP SPECT scan, the left parietal cortex adjacent to the nidus showed abnormally decreased perfusion. DIAMOX activated 123I-IMP SPECT scan showed perifocal hypoperfusion in the area within the left parietal lobe as well as the limitation of cerebral vasodilatory capacity in the perinidal area [Figure:2].

The patient was operated with standard microsurgical techniques. Through a subtemporal approach the main feeder from the posterior cerebral artery (PCA) was clipped. Through the interhemispheric approach, temporary clips were applied to the middle cerebral artery (MCA) and the anterior cerebral artery (ACA) feeders. The lesion was subsequently completely excised. Intraoperatively, the local cortical blood flow (lcoBF) was more than 20 ml/100 gm/min, when the temporary clip was placed on the main feeder from the MCA [Figure:3]. CT scan brain on postoperative Day 1 showed cerebral edema and an enlarged vascular tubular enhancement in the left parieto-occipital lobe, suggesting hyperemia [Figure:4] upper). Significant hyperperfusion was found in the left frontoparietal lobe in the dynamic scan image of 123I-IMP SPECT scan. However, on postoperative Day 1 and Day 8 static images of the0123 I-IMP SPECT scan study showed an area of hypoperfusion in and adjacent to the nidus site. This suggested that the hyperperfusion had disappeared on postoperative Day 8 [Figure:5]. CT scan also showed an improvement in the hyperemic state. CT scan done 3 weeks later showed marked reduction in the cerebral edema [Figure:4] Lower).

Postoperative angiogram done on Day 5 showed stagnation in the feeders and delayed circulation [Figure:6]. One month later, the angiograms showed normalization of flow in the stagnating arteries. There was no postoperative neurological deficit.

 Discussion



AVM surgery can be complicated by postoperative cerebral edema and hemorrhage in the adjacent brain tissue. Various theories have been put forward to explain the hemodynamic basis for this phenomenon, which include disordered autoregulation causing NPPB and obstruction of venous drainage leading to occlusive hyperemia.

1. Concept of normal perfusion pressure breakthrough (NPPB)

Normalized perfusion pressure in parts of vessels whose autoregulatory capacity has been lost following surgical resection of a large, high flow arteriovenous malformation (AVM) is thought to be a transitory cause of NPPB. Resumption of normal perfusion pressure in the brain around the AVM is believed to result in local capillary breakthrough, leading to uncontrollable cerebral swelling and hemorrhage.[1]

Al-Rodhan et al[2] presented an alternative concept of occlusive hyperemia. They argued that postoperative intracranial bleeding or edema may result in (1) occlusion of the draining venous system in the brain surrounding the AVM, followed by passive hyperemia and stagnation; and (2) stagnation in the feeding artery for the AVM and in the blood flow in the parenchymatous branching of the artery, followed by exacerbation of pre-existing hypoperfusion, ischemia, or edema.

Wilson et al[3] argue that this condition is observed frequently following embolization. Rapid neurological deterioration follows thrombus formation in a main draining vein. This is called “venous overload”. They state that venous overload can be “malignant” if venous occlusion occurs in the presence of nidus remnant.

2. Histopathological assessment of brain surrounding the AVM

Histopathological examination of AVM-surrounding tissues revealed strongly GFAP-positive gliosis in the region within 1 mm of the nidus of large AVMs; gliosis of one-to-two layers only in medium AVMs; and no gliosis around small AVMs. NF stain showed degenerative thickening of axons in the region surrounding the gliosis in cases of large AVMs and little change in axons around small AVMs. MBP stain showed lower stainability around large AVMs, indicating the presence of demyelinating change.

Sekhon et al[4] concluded that there was increased capillary density and absent foot processes in some vessels, which develop as a result of neovascularization in response to chronic cerebral ischemia. This anatomical configuration makes them prone to mechanical weakness and instability following the increase in perfusion pressure that occurs in adjacent brain parenchyma after AVM excision.

3. Prediction of NPPB occurrence

Factors indicative of NPPB occurrence include[5],[6],[7],[8],[9]: (1) Unsuccessful pre-operative vascular dilatation in the brain surrounding the nidus, coupled with (2) the presence of a hyperperfusion area confirmed by SPECT immediately after the operation, (3) resolution of the hyperperfusion within about 1 to 3 weeks after operation, and (4) increase in local CBF around the nidus after temporary clipping of the feeder, observed in intraoperative measurement of local CBF, being consistent with data from the SPECT performed immediately after the operation. These findings, along with cerebral swelling shown by postoperative CT and leaking of contrast medium suggest increased vascular permeability in patients with postoperative hyperperfusion in the surrounding brain. This may be easier to understand with the knowledge that autoregulation was already lost pre-operatively, as indicated by the examination of autoregulation by Diamox® loading. CBF around the AVM, a large one in particular, becomes chronically inadequate, leading to impaired autoregulation in the state of vasodilatation. AVM resection is thought to result in hyperemia due to postoperative rapid resumption of blood flow in the vasodilated vessel. A stagnating artery shown on postoperative angiogram supports this explanation. Consequently, intra- and postoperative minor mismanagement in patients with the above background, is likely to result in bleeding. However, in cases of middle-sized AVMs, marked stagnating arteries are noted among the main feeders, on a postoperative angiogram. Such stagnating arteries may be still seen even after CBF has lowered as indicated by SPECT. From the above, it is suspected that autoregulation in the surrounding brain is resumed in 1 week but the loss of autoregulation of the main feeder lasts for about 1 month. This difference was attributed to a difference in vascular response between the surrounding area, where passive vasodilatation occurs for maintaining autoregulation in the presence of steal-induced ischemia; and the main feeder, where active vasodilatation takes place for blood supply to the nidus itself, resulting in delayed vascular normalization and resumption of autoregulation in the latter case.

4. Is NPPB present?[10]

As resection of the nidus proceeds, a hyperemic state develops in the surrounding brain where autoregulation has been lost for an extended period of time, resulting in maximal dilation of expanded capillaries into which arterial blood directly flows, in the final phase of the operation. Inappropriate management of the feeder stump or hemostatic measures induce bleeding from the stumps of the arterioles or capillaries. If bleeding continues, it may lead to a vicious cycle of intracranial pressure elevation resulting in further abnormal bleeding, increasing the likelihood of NPPB occurrence.

Causes of intraoperative NPPB occurrence

In patients with a particularly intense steal phenomenon, expanded capillaries and small arteries are maximally dilated under disautoregulation in the surrounding brain, causing a hyperemic state. At this stage, breakthrough bleeding from the vessels of the impaired surrounding brain can be caused by faulty surgical strategies including (1) occlusion of the drainer in the presence of residual feeder, (2) inappropriate hemostatic operation for expanded capillaries into which arterial blood flows, i.e., insufficient time or intensity of coagulation or non-use of hemo-clip for vessels over 1 mm in diameter, and (3) intraoperative inappropriate blood pressure control.

5. Prevention of intraoperative NPPB occurrence[10]

Pre-operative endarterial embolization: (1) Feeder embolization and (2) nidus embolization are available as selective embolization of feeding arteries for NPPB prevention with the use of a microcatheter introduced into the vessel pre-operatively.

(1) Feeder embolization refers to the occlusion of feeding arteries not easily accessible with a remnant nidus and thus requiring a pre-operative session/sessions. In the case of proximal occlusion of a main feeder only, prevention of NPPB is non-significant because of increased hemodynamic stress on the small feeders in the surrounding area and elevated inflow into a non-occluded remnant nidus of low vascular resistance.

(2) Nidus embolization refers to an embolization extending from a feeding artery to the nidus. This technique is thought to accelerate a natural course of thrombogenesis by inducing stagnation of blood flow from other feeders via rapidly decreasing outflow into the drainers. This approach is expected to decrease the nidus in size and lower the A-V shunt volume, leading to resumption of autoregulation in the surrounding area and hence NPPB prevention. In patients who underwent an actual operation, the state of the surrounding brain did not change considerably when the operation was conducted 1-2 weeks after embolization, but in those subjected to operation at about 1 month after embolization, vasoparalysis of the surrounding brain improved, suggesting NPPB prevention.[1] On the other hand, embolization extending into the nidus involves risks such as venous occlusion by an embolizing material, resulting in increased intra-nidus pressure; or isolated nidus formation by an embolizing material migrated postoperatively, causing bleeding. Risk of venous occlusion is particularly high in cases of multifeeders coupled with a single drainer.

Important points in intraoperative bleeding control

The AVM nidus is excised in the following manner: (1) Temporary clipping is performed in as many feeders as possible; (2) The part of the nidus to be dissected is determined before excision is initiated; and (3) with at least one main drainer left, the nidus is gradually removed. Intraoperative bleeding is unlikely in an excision following the above procedure with repeated hemostatic measures taken as necessary. However, once mismanipulation occurs resulting in bleeding from the AVM, microvascular pressure increases making hemostasis difficult in many cases.

1) Management of feeders over 1 mm in diameter

As this size of the vessel indicates loss of autoregulation and lowered auto-constrictive capacity, mechanical hemostasis with the use of hemo clip will be needed following adequately long coagulation.

2) Management of expanded capillaries or moja moja vessels

These vessels are usually less than 20 mm in diameter, but as the operation proceeds the vessel dilates, eventually to over 200 mm in the later phase of the operation. The use of hemo clip is not possible in this case.

Coagulation is to be achieved underwater, by holding a vessel gently without closing the bipolar completely (wet-bipolar coagulation system) and by decreasing blood flow through mechanical pressure applied centrally. When these measures do not achieve satisfactory hemostasis, applying pressure to the bleeding site with compactly folded oxycel cotton moistened with Biobond® may work. If bleeding does not stop by applying pressure only, caution is required, because intracerebral hematoma may develop right below the compressed site. Coagulation must be ensured for expanded capillaries prior to complete occlusion of veins, and for vessels supplying the normal side of the brain.

Failure to achieve hemostasis at the capillary level: The source of hemorrhage should be located by tracking down the capillary to the arteriole level as and when required, to achieve complete arrest of bleeding.

Prediction and countermeasures for intra- and postoperative perifocal hyperemic state and NPPB occurrence.

Caution should be exercised for patients who show an intense steal phenomenon in the area surrounding the nidus, which is indicated by low perfusion detected by postoperative SPECT and decreased vasodilative reaction in the Diamox® loading test, and who show increased peripheral blood flow, by clipping of feeding arteries, as perfusion breakthrough-associated complications may develop postoperatively.[7],[10] In these patients who are presumably prone to developing a hyperemic state postoperatively, blood pressure should be maintained below 150 mmHg during and after the operation, and the barbiturate dose should be adjusted downward about a few days to one week after the operation.

 Conclusion



NPPB is responsible for massive hemorrhage and cerebral edema even after meticulous hemostasis during AVM surgery. SPECT study and local cerebral blood flow study should be performed. Preoperative 1) feeder embolization and 2) nidus embolization are available for selective embolization of feeding arteries for prevention of NPPB. Intraoperative bleeding control should be tackled by the management of feeders over 1mm in diameter, with hemo clips along with adequate capillary coagulation.

References

1Spetzler RF, Martin NA, Carter LP. Surgical management of large AVM by staged embolization and operative excision. J Neurosurg 1987;67:17-28.
2Al-Rodhan NRF, Sundt TM Jr, Piepgras DG, Nichols DA, Rufenacht D, Stevens LN:Occlusive hyperemia: a theory for the haemodynamic complications following resection of intracerebral arteriovenous malformations: J Neurosurg. 1993;78:167-76.
3Wilson CB, Hieshimia G. A new way to think about an old problem. J Neurosurg 1993;78:165-6.
4Sekhon LH, Morgan MK, Spence I. Normal perfusion pressure breakthrough: the role of capillaries. J Neurosurg 86:519-24.
5Awad IA, Magdinec M, Schubert A. Intracranial hypertension after resection of cerebral arteriovenous malformations. Predisposing factors and management strategy. Stroke 1994;25:611-20.
6Spetzler RF, Martin NA. A proposed grading system for arteriovenous malformations. J Neurosurg 1997;65:476-83.
7Kato Y, Sano H, Takeshita H, et al. AVM Has NPPB been resolved? Prediction of NPPB occurrence based on the surrounding brain at pre- and post-AVM excision and cerebral hemodynamics. Jpn Conf Surg Cereb Stroke 1996;24:421-30.
8Nakao K, Spetzler RF, Yamada K, et al. Immunohistochemical examination of tissues around cerebral arteriovenous malformations: With respect to causes of normal perfusion pressure breakthrough. Jpn Neurosurg Soc Abstr 1989;48:5.
9Sakaki T, Matsuyama T, Nakase H, et al. AVM : Has NPPB been resolved? Extent of intra- and post-operative involvement of NPPB in large AVM surgery. Jpn Conf Surg Cereb Stroke 1996;24:431-8.
10Yasui N, Suzuki A, Namide I, et al. AVM : Has NPPB been resolved? Problems in the treatment of large arteriovenous malformations. Surg Stroke 1996;24:439-45.
11Hashimoto N, Iwama T, Nishi S, et al. AVM: Has NPPB been resolved? Intra- and post-operative bleeding and cerebral swelling in AVM surgery. Jpn Conf Surg Cereb Stroke 1996;24:417-20.